On 21/08/2024 18:11, Darrick J. Wong wrote:

On Sat, Aug 17, 2024 at 09:48:00AM +0000, John Garry wrote:

quoted

For when an inode is enabled for atomic writes, set FMODE_CAN_ATOMIC_WRITE
flag. Only direct IO is currently supported, so check for that also.

We rely on the block layer to reject atomic writes which exceed the bdev
request_queue limits, so don't bother checking any such thing here.

Signed-off-by: John Garry <john.g.garry@oracle.com>
---
  fs/xfs/xfs_file.c | 14 ++++++++++++++
  1 file changed, 14 insertions(+)

diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 9b6530a4eb4a..3489d478809e 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c

@@ -1149,6 +1149,18 @@ xfs_file_remap_range(
  	return remapped > 0 ? remapped : ret;
  }
  
+static bool xfs_file_open_can_atomicwrite(
+	struct inode		*inode,
+	struct file		*file)
+{
+	struct xfs_inode	*ip = XFS_I(inode);
+
+	if (!(file->f_flags & O_DIRECT))
+		return false;
+
+	return xfs_inode_has_atomicwrites(ip);

...and here too.  I do like the shift to having an incore flag that
controls whether you get untorn write support or not.

Do you mean that add a new member to xfs_inode to record this? If yes, 
it sounds ok, but we need to maintain consistency (of that member) 
whenever anything which can affect it changes, which is always a bit 
painful.

John